← Moonshots with Peter Diamandis
Moonshots with Peter Diamandis
AI Insiders Reveal Elon Musk's Master Plan to Win AI w/ Dave Blundin & Alex Wissner-Gross | EP #192
01:50:55
Loading summary
Transcript252 lines
- [00:01]
A
XAI launches ROC code fast1. I had to double check the numbers on this because they were pretty epic.
- [00:07]
B
It's insane.
- [00:09]
A
Elon went from zero to building Colossus 1 in 122 days. Everyone said it couldn't be done. This guy will not slow down. He wants to be number one. And we're seeing these data centers leapfrogging each other.
- [00:22]
B
Elon, the entrepreneur of all entrepreneurs, knows that it's all or nothing. You don't build the second biggest data center, you either win the race or you don't win the race.
- [00:31]
C
This is the bitter lesson as applied to hardware scaling. It's sort of a case study in brute force hardware scaling where we're seeing the power, the chips, the data centers all being brute force scaled.
- [00:44]
A
These incredible tools are demonetizing and democratizing at an extraordinary rate all the students.
- [00:50]
B
Struggling with their largely irrelevant curriculum. This is what you should be doing.
- [00:56]
A
Now.
- [00:56]
B
That's a moonshot, ladies and gentlemen.
- [01:01]
A
Welcome, everybody, to Moonshots and our weekly episode of WTF just happened in tech. I'm here with my moonshot mates, Alex Wiesner. Gross. Dave Blunden. Salim Ismail is in India right now. We'll talk about that in a minute. But this is the news we hope that you watch that makes you optimistic about the future. That raises your IQ points 20 and gives you a chance to see the future. So, Dave and Alex. Good morning, guys.
- [01:28]
B
Good morning.
- [01:29]
A
So how was Labor Day for you?
- [01:32]
B
I got a lot of grief, actually, for being distracted and antisocial. I was talking to my AI agents. They were pestering me with progress all weekend. They have an IQ of 148 now, so it's pretty hard to deny them their. They're getting needy for the first time. You know, I've been waiting for this moment since I was 14 years old, so it's pretty hard for me to tune it out.
- [01:52]
A
I missed you, Dave.
- [01:53]
B
But they are getting needy.
- [01:55]
A
Alex, did you labor over Labor Day?
- [01:57]
C
Absolutely. Never a dull moment.
- [01:59]
A
Yeah. Yeah, me too. I'm in Port Townsend, Washington, near the San Juans. I have been here for a week with the family, which means I'm getting up super early, like at 5am just to actually get my email done and work and do my writing and then spend time with my 14 year olds on the beach, in the woods, fishing, went fishing, and I caught like a massive fish, which was about 6 inches high compared to Alaska. Anyway.
- [02:27]
B
Did you eat it? Just eat it raw right out of it.
- [02:30]
A
Catch and release. Catch and release. Okay, well, so listen, I Just want to take a second and just appreciate the fans that we've had on this podcast. It's been pretty amazing to get the feedback and you should all know we love doing this. We spend a lot of time working on this really to deliver the news that we think is the news worth learning about that gives you a positive view of the future. I want to take a second to read you guys some of the comments and just say thank you for the awesome comments to everybody. So why not? Jack says Moonshots is the best thing I've ever found. High praise. The Crypto canvas has best podcast and technology right now. Thank you guys for doing this consistently. And we do love doing it consistently. SteveDars1234 says an exciting future. Thank you for providing an optimistic long view amidst the constant doom and gloom of the news cycle. And I think that's one of the principles here, is if you are constantly watching all the negative news, it's going to shape your mindset. It's going to a really dystopian fashion. Carl Rankin 5385 says, quite simply, the very best and most relevant AI and digital technology podcast available today. Thank you, Peter, for allowing us to hear Selim, Dave and Alex and their collective brilliance. You're welcome, Carl. Renice IB6532 I absolutely love this podcast. You guys are doing a great job keeping up with everything. And yes, Alex is brilliant. Okay, let's follow up on that note from Polly Mewperl, who says Poly Merple. Go Poly Merple. Thank you. I was a little iffy on Alex at first. Then I realized I was just jealous of his intelligence. Now he's my favorite to see in the lab.
- [04:17]
B
That is exactly my experience with Alex when I first met him years ago.
- [04:22]
A
Oh, it's great. Sweetheart of a guy. And we'll wrap it up with Bill Jacobs, 33 86, who says the Fab Four are back.
- [04:32]
B
Fab Three.
- [04:33]
A
Well, today's the Fab Three. Yeah. Before we jump on to where Salim is, I just want to say thank you to our production team who've been amazing nixing Dana Khan and Gianluca Mangione. Thank you guys for all the hard work you do making this easy and fun. And we do have fun. I mean, it's pretty amazing. So Saleem right now is in India about to get on stage with his Singularity mates at an SU summit there. And before we hung up with him a few minutes ago, our edict to him was bring back a box of iPhone 17s. And please fix the US Indian trade issue. So he's taking that on. We'll see how he reports out.
- [05:21]
B
And this is Peter. This is back to school week. I don't know if you're in phase, but everyone is is back on campuses now, grinding away on the soon to be irrelevant curriculums that are, that are falling by the wayside. I got lots of questions from my kids, nieces and nephews over the weekend about what they should be doing, what they should be studying. How do you. And it's so great that we have Alex here to help add to that guidance because it's changing so quickly, very, very hard to keep up. I know that what they're learning is irrelevant and becoming more relevant by the minute. So we've got that much figured out. But then what is relevant and how are we going to keep up with it? So class starts on Thursday morning.
- [05:57]
A
Wow.
- [05:58]
B
Foundations of AI Ventures.
- [06:00]
A
My boys as well, but not quite. Not quite. MIT curriculum. They're eighth grade, but hey, that's good.
- [06:07]
B
Well, Alex, I don't know if you know, but MIT added a new thing this semester, 6e, remember you know, course six, which is computer science. And Double E had always had six A where you, you go to companies for a semester or two and learn how the real world works. They added 6e now, which is incredible. E is for entrepreneurship. So you basically take a couple semesters and go either work at a startup or a venture fund and see how the startup world works. And it's incredibly popular. So this semester I'll be teaching advanced algorithms in that curriculum a couple times this semester and then full time the following semester.
- [06:44]
A
I do believe, I do believe the career of the future is entrepreneurship, period. And we should have that conversation and we should talk about education on the next pod. We do go a little bit deep and there's some news developing there. Alex, I want to just not miss this point right now. If you're an incoming freshman to college, what's your recommendation? Skip it? Skip college or what do you do?
- [07:09]
C
It's a tricky time. I think it depends entirely on the freshman's goals. If your goal is to build a startup, I think there's a strong macroeconomic incentive to just do it. Now consider dropping out, moving to Silicon Valley, or doing it in Boston or elsewhere. But I think timelines, AI, AGI, ASI timelines are so short that almost any conventional career plan. If we had had this conversation 20, 30 years ago, I think it would have been far easier to project out sort of a conventional life plan or career plan? I think now the singular bit of advice, no pun intended, I'd have for any college freshman is assume that AI timelines are incredibly short. Assume that we're going to have superintelligence to the extent it doesn't exist somewhere already and just isn't evenly distributed. Assume that we're going to have superintelligence in the next two to three years, and guide your career plans accordingly.
- [08:09]
A
Yeah, I'll add my opinion there, which is, as I've said over and over again, find a problem you're passionate about, right? The technology is going to constantly change, but the problems are going to be fundamental for some time and then apply intelligence to that problem. You know, apply AI to problems that you care deeply about. So, you know, if you don't know your massive transformative purpose wherever you are, if you're in high school, if you're in college, if you're in graduate school, you know, pause what you're doing and really focus on what's your driver. You know, Mark Twain's favorite My favorite quote of Mark Twain, two important days in your life. The day you were born, the day you found out why, so why are you here? And then apply AI and digital superintelligence to that why?
- [08:53]
C
All right, shall we dive into the other? Peter of course, the other Mark Twain quote that's apropos to the college experience is the classics are books that everyone wants read, but no one wants to read, so maybe those two end up colliding.
- [09:08]
A
In this case, we have, we have Google, Ellen, you know, summaries. Now for us to be able to listen to those books in brief, in.
- [09:16]
D
A podcast every week, my team and I study the top 10 technology metatrends that will transform industries over the decade ahead. I cover trends ranging from humanoid robotics, AGI and quantum computing to transport energy, longevity, and more. There's no flow, only the most important stuff that matters that impacts our lives, our companies, and our careers. If you want me to share these metatrends with you, I write a newsletter twice a week, sending it out as a short 2 minute read via email. And if you want to discover the most important meta trends ten years before anyone else, this report's for you. Readers include founders and CEOs from the world's most disruptive companies and entrepreneurs building the world's most disruptive tech. It's not for you. If you don't want to be informed about what's coming, why it matters, and how you can benefit from it. To subscribe for free, go to dmanandis.com/metatrends to gain access to the trends 10 years before anyone else. All right, now back to this episode.
- [10:11]
A
All right, shall we dive into AI wars?
- [10:14]
C
Let's do it.
- [10:15]
B
All right, let's.
- [10:15]
A
Ready?
- [10:16]
C
Absolutely.
- [10:17]
A
Let's see, we're first. All right, first up is our friend Elon and his rollout of CO Colossal 2. Coming online in a couple Colossus 2 and coming online in a couple of weeks. One gigawatt data center in Memphis. And I love the fact that we're beginning to talk about these as energy, not number of GPUs again. And this is the mythical 1 gigawatt center is finally coming online, fitted for 500,000 Nvidia Blackwell GPUs and then doubling it again next year in 2026. And this is where Grok 5 will be born. Let's talk about it. I mean, this guy will not slow down. He wants to be number one. And we're seeing these data centers leapfrogging each other. Alex, what are your thoughts here?
- [11:06]
C
This is the bitter lesson as applied to hardware scaling. It's brute force. It's sort of a case study in brute force hardware scaling where we're seeing the power, as you mentioned, the chips, the data centers, all being, in the style of the bitter lesson, brute force scaled. What we haven't seen yet and will I think be very interesting to see, is that the same level of brute force efforts being applied to the software side of the stack. And I wouldn't be surprised to see that kick in as well. But for now, in the style of the bitter lesson, it's just absolutely incredible to watch with such vigor and such brute force, the hardware scaling side of the equation taking place.
- [11:50]
B
So the bitter lesson is from a guy named Richard Sutton who made this, I think, incredibly important observation that has permeated now the AI community. Alex, it'd be great if you give us a quick summary of it.
- [12:02]
C
Yeah, the core, I'll caricature it. But the core thesis behind Sutton's bitter lesson is that all of these decades by AI researchers spent developing artisanal solutions to problems, to speech recognition, to language understanding, to computer vision, were all basically wasted. And that in the end, all that really mattered was taking large data sets and lots of compute and off the shelf algorithms and just scaling them up to very large sizes. So this is the bitter lesson. This is why it's bitter. This is that all of this effort by humans, by human researchers getting their PhDs over the decades, coming up with artisanal New algorithms that they could publish to identify. You remember like several years ago even it was breaking news when researchers were able to build a cat detector. Like that was news.
- [12:59]
B
Now that was just five years ago, right?
- [13:01]
C
Yeah, a long time ago. And that's just been steamrolled by completely general algorithms with very little human injected prior knowledge, combined with huge amounts of data, huge amounts of compute. So the bitter lesson has been, I think one of the core themes of the AI revolution we've seen over the past few years. And going back to Colossus 2, the bitter lesson applies not just at the software layer, but also at the hardware layer.
- [13:27]
A
So some reads, brute force.
- [13:30]
C
Yes.
- [13:31]
A
Yeah. And what I want to talk about here is again just this leapfrogging of on data centers and on the hyperscalers. Elon went from zero to building Colossus 1 in 122 days. Everyone said it couldn't be done. Right now he is up against others and he's built the largest data center and is going to maintain that lead. We're going to see OpenAI coming out with their Stargate centers. So how will Stargate compare to Colossus?
- [14:07]
B
Toe to toe, actually. Isn't it right? They're exactly right in line. This footrace is very, very much winner take all. I know we had that debate last time we podcasted. It looks like there'll be five, seven big AI labs. But I think Elon, the entrepreneur of all entrepreneurs, knows that it's all or nothing. You don't build the second biggest data center. You either win the race or you don't win the race. Because these things, once they're trained, they compile down to something crazy fast and very easy to tailor into specific use cases. But nobody wants to start with the number two model or the number three model. And so you either win the race or you don't. And so Elon's all in, Stargate is all in. Sam Altman's all in. But yeah. Alex, do you know the exact numbers? I think they're pretty much right in a line.
- [14:50]
C
Yeah. Also, the numbers are a little bit tricky because I would distinguish between data centers for training, new models, and data centers that are going to be used primarily for inference. So the world, it would appear that we're moving toward is we're just tiling the earth's surface with inference time compute models that are.
- [15:06]
A
I think it's important just for a second for our viewers and listeners. Alex, talk about what's the difference between training and inference? If you haven't heard the Terminology before it's fundamental. Sure.
- [15:18]
C
So conventionally the way one would think about this is an AI model like ChatGPT is trained. It's created basically from large data sets at one time, sort of a fixed cost upfront, and then later on there's so called inference time when the model is actually used. If we were to sort of analogize this to software engineering, there's sort of compile time or development time when a computer program is created and then execution time when a computer program is actually run. Same idea here. So inference time when an AI model like the GPT series, for example, is actually deployed and being run in practice, those, those inference time use cases, based on the headlines that we're seeing, those are going to be run in data centers around the world. And I think we'll get to this later. Later we're standing up data centers as a human civilization all around the world, in the Middle east, in India, in Norway, elsewhere. But the training time, when the models are actually the frontier models, the strongest models that we have are being created. Those appear to be more geographically localized in the US at the moment. So it's sort of going back to Dave's question. It's a tricky distinction to distinguish between data centers that are going to be primarily intended for training and creating new strong AI models versus data centers that will be primarily intended for running models that already exist.
- [16:44]
A
I think the, another analogy here is if you go to school, learn a language, you'll spend, you know, a few years learning that language, but once you've learned it, actually speaking it, you know, is a lot quicker. And obviously inference is just that, having the language uploaded into your, your neocortex and being able to speak it. I think one other thing worth mentioning here is Elon's got big, basically access to infinite capital. Every time he goes to raise capital, he's over subscribed. Right. There is a massive amount of family office money, sovereign wealth money that's just prepared to fund his continued growth. It's never been that way. And I think what we're going to see here is the differentiator for the United States on building these companies, building these frontier models is access to this risk capital which doesn't exist at this level anywhere else. And I think that's pretty awesome.
- [17:36]
B
Yeah, Just to put some numbers on that, this is a million, a million Blackwells. They're $30,000 each. Remember a GB200 has two Blackwells on it. So I'm pretty sure he's talking about a million Blackwells, not a Million gb, two hundreds. But Alex, you probably know the answer. But anyway, a million Blackwells at $30,000 each, that's a $30 billion investment in the chips alone. Then know whatever. On top of that for the, for the, you know, the racks and the power supply and all that.
- [18:05]
C
It'S also probably worth just briefly mentioning that the electricity side supply of this so Colossus 1 sort of famously is has a self contained electricity source. It's using natural gas co generation facilities on prem it's not, my understanding, is not drawing electric power materially from the grid. It's generating on site its own electric power. And one can sort of extrapolate all sorts of interesting questions. Does this mean that there's going to be sort of a pocket economy of data centers that are being forced to co locate with nuclear power plants with natural gas cogen simply because the rest of the grid and the outside economy is too slow to catch up?
- [18:47]
A
Yeah, I think that's going to be the case. Right, where do you have cheap electricity? Just move your data centers there. All right, let's go on to the next one. Yeah, go ahead.
- [18:58]
B
So Daniela Roos and I took a tour of the Markly data center here. It's the first quantum deploy. But Jeff Markley, great guy who built the data center, he bought all the three megawatt generators in the country. What do you mean all of them? He said, well all the 5 megawatts were already sold out and I panicked because we need to generate a gigawatt. There are only so many of these generators. So now that's now the issue, like even if you get access to a power supply, you need the generators to turn it into electricity and those things are completely sold out.
- [19:31]
A
One of the things I want to talk about on the pod here is where would you make your next investments? Right, so we missed the intel, you know, option call we talked about last time with, with Leopold. I'm thinking, and this is not investment advice, this is me advice that you know, investing on the cutting edge of energy production. I mean just the drawdown right now, you know, as Eric Schmidt said to us, Dave, when we were having our podcast with him, you know, AI is energy limited, not not chip limited, not intelligence limited. It's energy limited.
- [20:06]
B
Well, I'll tell you what guys, why don't we go down on some podcast, maybe next time go down Leopold's holdings from his 13F filing and look at everything. Because he didn't just his biggest position is intel, but there's a whole list of things there that are all direct implications of what you just said. And so let's just analyze them one at a time.
- [20:22]
A
And you want to talk. You want to turn this into a financial investment podcast.
- [20:26]
B
Okay, we could label it. So if people want to skip that one, you know, they can.
- [20:30]
A
All right, all right. The next next topic we're discussion is keeping on the Elon theme here. Xai launches Grok Code Fast one. So I had to double check the numbers on this because they were pretty epic. So this is optimized for agent decoding. Right. You see here on this a graph of model performance per 1 million tokens and Grok code fast one. I'm not sure that rolls off the tongue as a name, just trounces everything. I put the numbers down at the bottom here. So input tokens are $0.20 per million tokens, output a buck 50 per million compared to GPT5. Right. Input is $1 in the quarter compared to 0.2 and Claude Sonnet 4 input is $3. So we're talking about 15 times cheaper on input tokens and we're talking about 10 times cheaper on output tokens. How do you compete against that? I mean, this is a race to the bottom. Thoughts, Dave?
- [21:31]
B
It's definitely not a race to the bottom, even though it appears to be. This is a get the market share, don't lose, no matter what. People get addicted to this stuff so quickly and then they want an infinite supply of it. So this is much more like a crack dealer giving out the first hit for free than it is like a race to the bottom. And I think people are completely misinterpreting whether AI is a race to the bottom and also whether the chips will commoditize. Neither is going to happen because the demand is infinite.
- [21:59]
A
Yeah. Alex, your thoughts here, buddy?
- [22:01]
C
Yeah. What's worth, I think, noting is if you try to interact With Quantum Grok Codefast1 via browser, you will not find it. And we've talked on the POD in the past about the browser wars, about browsers, web browsers as distribution channels for AI. I think it's quite notable, but sort of under the category of burying the lead that you can only access as a consumer, you can only Access Grok code FAST1 via one of several different coding environments like Cursor or Windsurf. So to the extent that we talk about distribution channels for superintelligence, I think it's actually quite notable that those coding environments are becoming almost competitors for the browser for accessing superintelligence.
- [22:45]
A
Fascinating. Fascinating. Yeah. Entry points. Yeah.
- [22:50]
B
So I'm looking at cursor right now on my screen. I don't see it there. Is there something I need to do.
- [22:55]
C
To get it with Windsurf? So it's a bit of a hassle to get to with Windsurf. You have to search specifically for Grok code fast one in order to get it. It's not even one of the recommended ones.
- [23:07]
B
Oh, no, no, it's here. It's here. Sorry. It's way down at the bottom. I don't know why they.
- [23:11]
A
Well, it's just out. It's just out and it'll gain popularity. But I think the point here, I mean, for everybody is these incredible tools are demonetizing and democratizing at an extraordinary rate. Right. And we're going to see literally billions of coders. Everybody will learn to code. The language of coding is going to be basically your mind, your. Your English or your Hindu or whatever it might be.
- [23:39]
B
All right, before the next podcast, I'll test it and see how it fares. Just on quality and obviously on price. It's incredible, but just on quality. But I don't understand how this price point is possible. Just. Just counting up the flops and working it back to the chip costs. Alex, I don't know if you have any insights on that.
- [23:54]
A
There was an article that was said that a lot of this is being basically carried by Silicon Valley investment, at least.
- [24:02]
B
Yeah, I mean, you can do a little bit of that, but you can't just burn money by the billion. I mean, you can for a little while, but you can't sustainably. You think that's what's going on here, Alex? I don't know.
- [24:15]
C
It's not clear. I see different accounting schemes. Without having direct access to the chart of accounts for the Frontier Labs, difficult to know whether inference time, as we were discussing, is profitable or not. I've seen claims either way, but I do think more broadly, Jevons paradox, broadly speaking, that the cheaper given commodity goes, that there's sort of a paradoxical net increase in demand resulting in potentially greater expenditures. In this case, I think we're going to see that with code generation, as well as the cost of intelligent code generation trends towards zero, as it has been for the past few years. I think we will see to Peter's point, we're just going to see just in time, code demand for everything, and we'll be awash in new code that otherwise never would have existed.
- [25:03]
B
Well, one thing that came up when we were talking to Kevin Weild 2 weeks ago @ OpenAI the Chief Product Officer at OpenAI is that there's a huge amount of routing optimization going on and when I use these things I'll bounce back and forth between a trivial question and then can you solve cold fusion back to back? And the models are very intelligent now about routing it to the minimal model that will actually answer the question correctly. So a huge amount of savings from those types of optimizations and there are many, many different layers of that. So I think you're going to see the innovation at this ridiculous pace and then the price point continues to come down. But again the demand is infinite, so.
- [25:42]
A
It won't let us know when you solve zero point Energy. I'd like one. Okay, okay, this is another fun article again. Elon verse Musk poaches 14 meta AI engineers with a different offer. So we'll talk about what's going on in meta in a second. But I think this is fundamentally what Elon does extraordinarily well. He has a massive transformative purpose, you know, open up Mars for humanity, make humanity multi planetary species. And he has that as a pure signal. And entrepreneurs who are willing to work hard, who are builders, want that. They want to work on something epic. It isn't about the money since eventually we're heading towards a post capitalist society anyway. So Elon's offering purpose and equity over cash and his equity has done incredibly well. I don't think he's ever started a company that's lost money. It's increased. All the companies from Starlink to Xai to X have basically just skyrocketed in value and we're going to see a startup intensity here. Alex, what are your thoughts on this?
- [26:58]
C
I remember back to the early days of, call it the post 2012 Imagenet revolution in AI when there was a lot of concern, including as I recall from Elon, that AI would end up being a monoculture. That one lab maybe called Google DeepMind would sort of completely capture the future. Lightcone with AI. And I think I view this and other related headlines as very healthful sign that we're not ending up in that future of, of a monoculture, of a singleton where just a single culture from a single frontier lab has complete dominance, complete hegemony over AI. We're going to see, and we are seeing multiple competing AI frontier labs with different cultures. And this should and is, you know, this should be one of those cultures. The manic focus and intensity of, of just delivering state of the art results. And we'll see lots of other cultures as well. And we'll have them compete. And that's the world that we humans, I would argue, want to live in.
- [28:06]
A
Yeah, and I think, you know, purpose, a purpose driven life is going to be far more important in the future than anything else. And being clear about why you're doing what you do and waking up in the morning and having an epic mission and being excited about building is a future that I want. You know, it's always been part of my life and I want for my kids. And that's what's going to win over just cash, especially working someplace. If you don't like the culture, as you said, I want to play a short video. This comes as part of Elon's Master Plan, Part 4 on Sustainable Abundance and let's chat about it afterwards.
- [28:45]
C
Humans are toolmakers, and at Tesla, we are builders of physical products at scale that make life better for all. We are building the products and services that bring AI into the physical world. We are combining our manufacturing capabilities with our autonomous prowess to accelerate global prosperity. We are building a safer, cleaner and more enjoyable world for all. We call this Sustainable Abundance.
- [29:30]
A
What's the happiest future you could imagine?
- [29:33]
C
One which there's Sustainable Abundance World.
- [29:36]
A
So I love that term, sustainable abundance. You know, it's amazing to see abundance becoming an underlying theme for the hyperscalers and for the tech world. And it's part of the optimistic vision of the future. Right. You're not going to see this on the 6 o' clock or 7 o' clock news. You're not going to see it on the Washington Post, the New York Times. It's important to realize that technologies that we're talking about on this pod are going to shape every aspect of our lives. And there is so much positive news, it's so easy to focus on the negative, but, you know, understanding and this is, you know, Elon's been on this mission since I met him back in 2000 when he'd sold, you know, sold to PayPal, sold PayPal to what you call it, eBay. Sold to eBay. And he's been on the mission of autonomous vehicles, electric cars, solar, you know, making humanity multi planetary. And people love those epic grand challenges, right? These moonshots, which his podcast is all about, and they gravitate towards that. Dave.
- [30:43]
B
So, Peter, you invented Xprize and used X about 10 years before Elon stole it from you. And then Abundance was the name of one of your books.
- [30:51]
A
My first book, yeah.
- [30:52]
B
So what's next in the Elon takes all. Peter? Ideas.
- [30:56]
A
Oh, God, it was very funny. Because the X Prize logo looked identical to SpaceX's and then to X's logo. I didn't have the heart to call him out on it, but hey, it's fine. He's been generous supporting the xprize and supporting our work over the years. He's given probably $150 million of capital to support some of the X prizes we've done. But sustainable abundance is a real thing. It's about digitizing, dematerializing, demonetizing and democratizing everything. The way I like to describe this is if you think about it, when Google came out, Google was, you know, a for profit company that had the biggest nonprofit impact. In other words, it uplifted all of humanity and the poorest child on the planet using Google and the wealthiest child, you know, Larry Page's kids, Sergey Brin's kids, Google was identical for those. It was a leveling and democratizing capability. And we're going to see that here for food, water, energy, healthcare, education. And that's extraordinary. Alex, you've been on this journey with me.
- [32:06]
C
Oh, yes. Yeah. No, I love this term. And I think it's sort of an implicit recognition that another important part of the equation, which is autonomy, AKA superintelligence or post, abundant intelligence, is sort of the missing factor here. So when I look at the video and I hear the term and I see its usage in practice immediately this screams to me the intersection of abundant energy and materials on the one hand and abundant intelligence and autonomy on the other hand. And I think it's sort of a, it's a rare and precious time in human history when we can have near total clarity as to what the technology tree looks like.
- [32:50]
A
Every time, every time we talk about having scarcity of something like lithium, oh my God, we're running out of lithium. It's like we discover these massive supplies off the coast of California. There is nothing that's truly scarce, period. And I think the sooner people get that technology is a scarcity destroying force. Dave, you were going to say something.
- [33:12]
B
Well, yeah, the other thing on that slide that is kind of sandwiched in the middle and bullet too is the seven day work week. You have a bunch of companies at Lynx Studio working 997 and they just declare we're doing 997. What's 997? It's this China thing where you work 9am to 9pm seven days a week while you're sprinting towards some massively transformative purpose. So those two things go hand in hand. Like no one's going to work 997 on something irrelevant. Right. It has to be something world changing and imminent. You know, 997 is no way to live your life. You know, you do it in a sprint to get to a very specific destination. But that is by far the winning strategy. Yeah, know if you, if you do this massive million GPU data center and you do it a year late, it's worth zero. It's absolutely worthless. There's no point in doing it, you know, on a four day Europe work week, you just either do it or don't do it. And if you do it, you sprint. It's like an Olympic gold medal. You gotta sprint.
- [34:07]
A
Yeah, yeah.
- [34:08]
C
I think it's also interesting when we. I think it's 996.
- [34:12]
A
Yeah, China was 996, but I think Davis actually.
- [34:16]
B
Oh, the slide said 7. Sorry.
- [34:19]
C
With the reaction to 996 popularly lying flat people who are opting out of the 996 culture, it's sort of interesting to think about whether sustainable abundance actually obviates that entire discussion altogether. Rather than lying flat under the implicit assumption that 996 and that the need for enormous amounts of human labor are going to continue in perpetuity, what if we actually, a few years from now, find ourselves in a sustainably abundant future where the need for 996 human labor is actually only a short term need. A few years from now, we hand over that workload to autonomous systems.
- [34:55]
A
Then we go to the stars. Then we go to the stars. All right. On the flip side of this conversation is people are beginning to bolt from Meta's new superintelligence lab. So two months after the launch, at least three top researchers have resigned. Two returned to OpenAI. While Meta's long term product director also joined Altman's ranks. Departures are raising questions despite recruits being offered nine figure pay packages. And to be clear, we don't actually know what's going inside. This is just reporting what came out and wired. But I do think when you're capturing an employee by offering them a lot of money, that's not going to capture their time and attention and their heart. It's mission and purpose that captures them. Dave, you agree?
- [35:43]
B
Well, absolutely. And I think also Elon has a reputation for everything, always working. So having an MTP has to be married with a mission that will succeed to develop that reputation of not only is it massive in implications, but it's going to happen. It's real because it's very easy to go out there. And say, hey, I'm going to build electric cars. Come on. And then if it doesn't work out, no one will join you the second time. So you have to have a track record of succeeding. Now, Mark Zuckerberg is probably the, you know, of the young CEOs in the country, the most successful with the most cash flow, every opportunity to win. Elon has done it repeatedly though, so I think a lot of people are flocking to the he's always been right before. Why would he be wrong this time? Yeah, hashtag that really helps.
- [36:30]
A
Never doubt Elon, right? Yeah.
- [36:31]
B
Yeah.
- [36:32]
A
All right. I want to spend too much time on this, but it's been a big week for Google. We're on the verge of the release of Gemini 3, which will be coming out any day now. But I mean, top news and extraordinary conversation around Nano Banana Powered by Gemini 2.5 flash image let's take a look at this video.
- [36:53]
E
A few days ago, image editing changed forever. Google released Gemini Flash 2.5 image. Nano Banana that has everyone buying puts on Adobe because Photoshop is officially dead. Instead of learning how to use all these antique tools, you can now just prompt nanobanana for changes and it's able to deliver any photo alterations you can imagine. And most importantly, while maintaining the consistency of the original image. Not only is nanobanana an exceptional image model that's already at the top of the LM arena leaderboard, but it's also extremely fast and affordable, costing only 3.9 cents per image via the API. The upgrade that most people are talking about though, is character consistency. If you start with an image of a person or pet, for example, the model can blend it it with a different image or make minor changes to it without noticeably altering the original character or multiple characters and objects, like this guy did, by blending 13 different images together. What's kind of crazy about this model, though, is that it also has an understanding of the real world. Like if you point to a spot on Google Maps and ask what a person would see there, it can generate a realistic photo.
- [37:53]
A
Just epic.
- [37:55]
B
Just epic is right. And very few people. Basically nobody under the age of 40 remembers life before the GUI, the GUI on your computer interface. But we do. When the Apple II first came out, you would boot it up, or TRS 80 or whatever, you'd boot it up and this little flashing prompt would be there and what can I do with this thing? Yeah, hello, what do I do? And all you can do is just start writing code. That's basically all you've got. Just start Writing code. And then what, 1984, 85, Steve Jobs comes out with the Mac and, and now everybody's lived in this kind of world of stasis of the GUI for 30, 40 years now. So everybody's like, yeah, nothing ever changes. This is all going to change imminently. It'll be the biggest step function. It'll be much bigger than going from no computer to computer or computer to command line to gui. But it's been so long that nothing has changed that people are completely underestimating how different the world will be a year from today when everything has a. I just asked the computer to do this for me, just like on Star Trek. And it just did it. But that's happening literally right now. The implications for startups are incredible. Like if Adobe gets destroyed by this, you know, I think our friend Greg Bellows is still over there. So it'd be kind of sad if that happens. But it'll be very important as a wake up call that if you've been camping on your software installed base for the last 20 plus years, milking it for money, your days are numbered.
- [39:27]
A
Yeah. I mean, because you're just.
- [39:28]
B
Everything's going to change if you are.
- [39:30]
A
Adobe Photoshop or a Canva, you know, specialist making your living that way. Right. I mean, you understood how to do layers and masks and manual adjustments, but Nano Banana is just, you know, edit through language, not layers. It like, you know, do this, you know, it's, it's literally, you know, how do you describe what you want in a way that AI will understand it? That's going to be the skill base.
- [39:56]
B
Yeah, exactly. And if you're a graphic artist or you're a writer or whatever, you get so used to these tools and all their proprietary interface components and then you're just afraid to shift to something else because it's so, you get so invested in knowing where the menus are and knowing where the buttons are, knowing how it responds. And so then you're locked in and you end up paying for that product for, you know, 10, 20 years. Now everything's wide open again. The interface is trivial. My mom, who's in her 80s, has no problem creating images. She could never use Adobe Photoshop, never, never figured it out. Now she can just talk to it.
- [40:29]
C
I would argue, actually. So I've been using nanobanana quite a bit. It's actually a much bigger deal than just some of the headlines that would say this is a Photoshop killer. So in using nanobanana, some of the most striking new capabilities that I'VE seen are you can feed it an image and then ask to view the same scene from a different perspective. That's way more than just pixel level Photoshop style editing. It smells to me like this is just a sliver or a distillation of a larger world model. Sort of like We've spoken about Genie 3 in the past. It feels to me like this is some sort of like tendril from a much more monstrous model. And if that is indeed the case, and to the extent that Nano Banana has basically become merged into mainline Gemini model releases by by Google DeepMind, I think this sort of portends a future where world models like in the class of Genie 3, the videos that we've discussed previously, those just merge into GPT or Gemini type models as the ultimate modality of interactive, simulated, maybe even streaming realities.
- [41:40]
B
I'll tell you what else Alex. All media competes with all other media. There's no swim lanes. Everything competes for time from users.
- [41:47]
A
Time is a great scarcity, at least.
- [41:49]
C
While we have While we're stuck with finite attention, maybe we can make attention post scarce as well.
- [41:53]
B
Okay, well for the next couple of years but you know, this time of year normally by now I would have, you know, done a fantasy football league signed up for my players know who you know, first football games will start. I haven't even paid attention. I don't even know what's going on because the stuff that you post on our link chat is so much more entertaining and engaging than mainstream media. The stuff that you as a single handed person can create is so much more interesting and relevant. It's just a capability that never would have existed a year ago.
- [42:22]
C
Well look, a singularity probably only comes about approximately once per planet. So it's a special time.
- [42:29]
A
It is a special time. We're going to celebrate that. I'm going to have a singularity party when it happens. I hope you guys will join me. We made this point last time. AI won't take your job. It'll let you do any job. And I think this is a perfect example, right? I mean literally, designers who have made their career based on understanding how to use a specific tool really well now anybody can do that. But here's a question for you, right? If I go and I say to nanobananda, hey, place me on the moon in a spacesuit, getting into a starship for return flight to Earth and it generates that. Who's the author of that? Is it the software? Is it the human who prompted it? We're going to start to have some interesting Conversations around ownership, blurring the lines of authorship and human in the loop creativity. So that's going to be an important conversation. But let's talk about the real issue here, which is the ability for this to drive accelerated misinformation and the erosion of visual trust. The old saying, seeing is believing is out the window, period. Thoughts?
- [43:44]
C
Yeah, my comment on that would be we're in a post vo3, post sora, post natural language generation era. I think there's a future to the extent that one wants to have faith in the accuracy of any visual inputs, images or videos. I can see a future maybe where there's some sort of cryptographic chain of trust between cameras, video and still cameras and browsers. Sort of like the way there's cryptographic guarantee that when you put in your credit card information to pay for something on a website, that the credit card information is handled in a cryptographically safe way between you and the ultimate counterparty. One can imagine some sort of cryptographic guarantee that the image that you see in social media was actually unaltered in some sense from the original capture without any AI involved. That said, my baseline expectation is that that's not going to be very popular.
- [44:46]
A
And blockchain to the rescue. But this is deep fakes on an industrial scale, right? I mean, just to put it where it is, nanocast.
- [44:56]
C
This is what people also said though, right before GPT2 and GPT3. And this will empower all, all sorts of misinformation, disinformation. And yes, there is a lot of probably false information that's being generated by these models. But I have to look at it as sort of a risk, reward, trade off. There's so much new scientific information that's being unlocked by these models. Very difficult to get too bothered by the potential downsides.
- [45:24]
A
I'm not worried about that, but I'm just saying for the majority of 8 billion people on the planet, if they keep on seeing, I mean, you're going to get to a point where when I'm there now, when I see a video, my first reaction is it real? Right? And my second reaction is going to be no, it's not real.
- [45:41]
B
Yeah, no, no, there's no doubt. You know, you've read all of the Neal Stephenson books, I'm sure, Peter. Of course, of course, of course it's not enough course, but of course. So yeah, like the diamond age to me was the. Everything he's ever predicted in those books has happened. And he invented the word avatar in the first book, Snow Crash just foresaw cyber cyberspace, that word was invented there. But in the diamond age, everybody moves into these communities that have different rules around technology and how you manifest it because it gets too weird too fast. And so the big, big step function change for society is cameras everywhere. And that started years ago. And so now we're living in the cameras every world. There's no concept of privacy. Anytime you're outside, you're being filmed at a minimum by a satellite.
- [46:30]
A
But you can know anything you want anytime, want, anywhere you want. The data is there. Layers of drones, layers of satellites, layer of autonomous car cameras everywhere. Everything's being imaged. There is no privacy. That's another conversation we can have.
- [46:44]
B
Well those, those 10 megapixel cameras now are $0.50 each.
- [46:48]
A
Yeah.
- [46:48]
B
So they're going to be everywhere. And so then now, you know, that already happened to society and also social media happened to society. Just totally changed the whole election process. And everything's, you know, everything's been disrupted tremendously in the last 10 years. So now you layer the deep fakes on top of that. It's just the third act in this massive, turbulent societal change that's only going to accelerate. And you know, of course governments do nothing. They just sit there and assume people are talking about the next election like it's going to be anything like the last election. It's going to be a different world by the time we get to the next election.
- [47:20]
C
Yeah, I would just maybe add, I would speculate that this maybe call it a moral panic that we're engaging in right now is going to look very quaint in a few years. It's difficult to imagine people with smart glasses doing real time augmented reality overlays of everything that they're seeing on the one hand, which by the way means basically everything becomes photo edited. Everything that you see becomes doctored by default. On the one hand. On the other hand. Oh, but you know, clutch, clutch, whatever it is the moral panic that you're worrying about, like what about the photo editing? No, I think it's far more likely this will look quaint and nonsensical in.
- [47:59]
A
A few years when everything is panic is becoming quaint. It's worth noting that there is a digital invisible watermark that Google's putting on these images. Synth id and we're going to start to have sort of a arms race between gen AI and detection tools as well. That's going to be part of it. It's always the, you know, the virus antivirus war, everybody.
- [48:22]
D
There's not a week that goes by when I don't get the strangest of compliments. Someone will stop me and say, peter, you've got such nice skin. Honestly, I never thought, especially at age 64, I'd be hearing anyone say that I have great skin. And honestly, I can't take any credit. I use an amazing product called OneSkin OSO one twice a day, every day. The company was built by four brilliant PhD women who have identified a 10amino acid peptide that effectively reverses the age of your skin. I love it. And like I say, I use it every day. Twice a day. There you have it. That's my secret. You go to Oneskin Co and write Peter at checkout for a discount on the same product I use. Okay, now back to the episode.
- [49:04]
A
Let's go to Google's next big announcement of this past week, which is Google Translate and another incredible product coming out right now. So Google's AI powered live translation. Historically, Google has translated about a trillion words per month for 600 million users, supporting 243 languages, which by the way is 58,806 language pairs. Amazing. But we're now being driven by Gemini 2.5 with a live translate. Let's take a look at this video.
- [49:38]
C
Hi there.
- [49:38]
B
My friend told me there's a sandwich.
- [49:40]
C
Here that's really good, but I'm not.
- [49:41]
B
Sure which one it is. It's spicy, has really tasty cheese on it and avocados.
- [49:45]
C
I think. Hola. I think I know what it is.
- [50:08]
B
It's seasonal. In fact, we've already taken it off.
- [50:09]
C
The menu, but let me see if we can still prepare it for you.
- [50:12]
A
Pretty amazing. The question is, what? What's this going to do to the industry of language translation? What's it going to do to people learning languages? You know, I used to want my kids to learn multiple languages. Now the question is, do they invest their time in doing that?
- [50:30]
B
Gentlemen, what was that company that all the kids used to cheat on their homework? They got a public company, got obliterated. Alex, which one is that?
- [50:41]
C
Not sure. But what I would say, I mean, just historic reminder. Remember the transformer architecture that sort of helped to kickstart a lot of the generative AI revolution? It was originally developed for language translation, for machine translation. It was an encoder plus decoder architecture. Right now we mostly use the decoder part. But nonetheless, it's sort of ironic that the original targeted application for Transformers was statistical machine translation or machine translation. And it's sort of only now that we're starting to see pervasive Machine translation, finally tackling the real world use cases of real time conversational embeddings. That's first thought. Second thought is it's sort of interesting to speculate, I think, what does this do to language diversity in general? Is this sort of a net promoter of languages? There's been a lot of hand wringing over the past 20 years about low resource languages dying out in favor of usually English, but sometimes other languages. Or is this sort of a net promoter of diversity where once all languages, thanks to AI, become fully interoperable, as it were, there's suddenly no reason to collapse down to sort of one modal English language?
- [52:04]
A
Do you know the joke here? What do you call someone who speaks three languages? Is trilingual someone who speaks two languages bilingual? If they speak one language, they're American. I love that.
- [52:15]
B
Well, that's going to turn out to be the winning strategy. What do you know? The company was Chegg. Check out its stock ticker, or we'll splice it into the podcast here. But it went down from 90 bucks to a buck 40.
- [52:26]
A
Wow.
- [52:27]
B
You know, just, just because, you know, ChatGPT is a better way to cheat on your homework or whatever. I'm. Well, that's not really what they do. But, but it's, it's that.
- [52:36]
A
But the point here is, the point here is that Google is also providing a language practice mode that allows you to personalize speaking and listening exercises. Right. And so the impact on that on Duolingo was a 10% stock drop. So Duolingo is a $13 billion company. It's done incredibly well. 130 million active users, only 10% of the users pay. But nonetheless, it's generating real revenues. And you can see this drop that occurred last on August 29th when this Google Live AI translate capability was announced. So this is another example where these large frontier models sort of in their wake, whether or not they know it, are going to be disrupting existing companies who are going to have to constantly be pivoting.
- [53:29]
B
Yeah, Alex, Duolingo is absolutely doomed unless it becomes an AI company. And if it becomes an AI company, it can go through the roof. But you know, a lot of these companies don't have the AI talent to get started. So you gotta, you gotta turn the battleship somehow. But if you do succeed in turning the battleship, your valuation can go through the roof. You've seen that a single incredibly talented AI researcher can be worth a billion dollars. So, you know, the value is there.
- [53:54]
C
I mean, in general, in the short term, I want to generalize from Just this one instance. So in some sense, I think the sort of the cliche here is every software as a service company is under existential threat from generative AI models that will simply cannibalize them from below. Whether you know you're doing software for some enterprise purpose or whether you're, you're just like software subscriptions to help people learn new language, that frontier model is just that Chatbot is going to, to devour you because you've become just one special case among countless, numerous cases that a generalist model can handle.
- [54:32]
A
I think that's critically important, right? Every CEO out there, every board of directors needs to understand that if this, they're not building on an AI base that's accelerating alongside with everybody else. If they're depending on their old business model, software as a service, they will be marginalized.
- [54:51]
C
But here's, here's.
- [54:53]
B
Remember what we said before too. If you're in a regulated industry, you have a little bit of time, you can actually get ahead of it. You got to get the AI talent now, but you can get ahead of it. If you're not in a regulated industry like you know, Chegg or Duolingo and you're just user install based, then you're really vulnerable.
- [55:07]
A
I mean you have an advantage of a user installed base and a brand. Use that to your advantage to actually leapfrog forward. Don't like yesterday idle? Yeah, yeah.
- [55:17]
C
Just to dwell for a minute on what that leapfrogging looks like. I think I agree. In the short to medium term, differentiated user experiences are a bit of a moat, if you will. But in the medium to long term, what I'd like to see from every single SaaS that that feels existential risk from being devoured by a generalist model is step up your ambition by 100x1000x. If you're duolingo and you happen to feel existentially threatened by generalist models, maybe consider becoming a brain computer interface company. Wouldn't it be wonderful if we could sideload new languages in the style of the movie the Matrix into the human brain? And rather than spending days or weeks or years learning a new language, why can't you enable your clients, your users to learn it in a minute? Yes.
- [56:04]
A
Pick a moonshot. I mean that's the whole purpose of this podcast. Get people or to go 10x100x bigger.
- [56:10]
B
Pick your moonshot or hire Alex as a consultant for two weeks and you'll have a moonshot at the end of that.
- [56:15]
A
No, there's no time for that. Okay. All right, so let's take a look here. We've got OpenAI Real Time API bringing smarter voice AI. So let's look at this quick video here. I love this one. Are there any homes in my budget? Neither.
- [56:33]
C
Water with a view of the skyline and Mount Rainier?
- [56:37]
A
Sure, let me look.
- [56:39]
C
With your buyability of 824k, Wallingford would.
- [56:43]
A
Be a great fit.
- [56:44]
C
I think you'll love 404 N. 33rd St. It has those skyline and Rainier views you're after.
- [56:51]
A
With this week's market, I'd book a.
- [56:53]
C
Tour with an agent soon. Want me to set that up?
- [56:56]
A
So this, you know, this capability, this is an example of using this on Zillow to find your home, describing exactly what you want and having it actually scrape and generate an efficient answer. But just the ability to do all of this and actually get you to the point. I mean, what you want next is find the house, buy it for me, arrange the mortgage and arrange the moving trucks and let me know when to show up in my new place.
- [57:24]
B
Yeah, no, I think AI as a management tool, a general purpose management tool, hugely underrated because everybody loves the graphical stuff, the image creation and the self driving car, the stuff you can feel. But just as a general way to manage large scale projects with hundreds of people and moving parts and logistics, it's unbelievably good at doing that. So I think we can expect far more efficient construction, management, manufacturing supply chains than we've ever seen before. Because, you know, the sensor data with all the cameras everywhere has been available for a few years now, but it's all kind of dumped into big databases. You throw it into snowflake or something like that and then very hard to make sense of it. The missing ingredient was this AI overlay that can just take the unstructured freeform data and turn it into conclusions, actions, schedules, buying things, scheduling things, managing things. You know, we had our, our condo in Vermont built many years ago, 20 years ago, they put the Tyvek on upside down so it's overlapping the wrong way. So it grabs rainwater and funnels it into the wood. So you know, like, you know, years later, everything's rotting, the whole thing's falling apart. Like why, why would you put the Tyvek on upside down now? Very, very easy for the AI to say, hey dude, stop. It's just as easy to put it on, right? You're putting it on overlapping the wrong way. Just a trivially easy AI problem. All of a sudden thousands of things like that can suddenly Be converted.
- [58:51]
A
Yeah.
- [58:53]
C
Oh, sorry, Peter.
- [58:54]
A
I was going to say the point here though is the real time API. Alex, let's chat about that.
- [58:59]
C
Yeah, so I've played with this. So the underlying model is called GPT Real Time. And if you've played with AVM Advanced Voice mode of OpenAI, that's the mode where you can chat in real time with very low latency responses. With ChatGPT, it's a lot like that, but in API form, so that it can serve as a backend for third party applications. And I really do think this is transformative in part because imagine taking sort of low latency voice to voice, but generally capable intelligence and now embedding it everywhere. I think it probably ends up being transformative for customer service type applications, probably many other sectors as well. But even bigger picture, I think this is a preview, albeit a tiny preview, of a future where every single audio segment, every single pixel on screen is generated in real time, streamed interactively on demand. And just our user experiences, our user interfaces are just completely, just in time generated. It's going to be a very, very interesting future.
- [60:08]
A
And there's a single interface. There's a single interface to the world, right? Yes. Your Jarvis will go and interface with everything out there, whether you know it exists or not, and give you the answer you finally want.
- [60:22]
C
We don't have a good catchphrase for this.
- [60:25]
B
Our voice customer service company vokara, doubled in ARR during the past week and is planning to 10x between here and the end of the year just using this exact capability for complex customer service and sales conversations. But so far the consumers dramatically prefer it to. A human call center agent is vocal.
- [60:46]
A
Because it's so knowledgeable in the Lynx studio.
- [60:49]
B
Yeah, yeah.
- [60:50]
C
To MIT team, To my knowledge, we're missing a term for this. I've definitely come around to the view that it's important to coin new terms whenever there's sort of this important new concept. I think we're missing a term for this. It's not conversational user interface, because it isn't always conversational. The best term, if I had to coin a term for what I think we're seeing the beginnings of, would be something like streaming interactive models. It's not necessarily just voice. Could be like Genie 3, where if there's a visual component could ultimately be like a brain computer interface type component. So try it on for size. Streaming interactive models or sims.
- [61:29]
A
And because everything becomes a tla, Streaming interactive models are sims.
- [61:35]
C
Right.
- [61:35]
A
Okay, Alex, you pointed here.
- [61:37]
B
Yeah. The Zillow example is really Important for people to look at, you know, rewind the pod and watch it again. Because customer service is typically a phone call today, it's very hard to explain complicated things on a phone call. So this is very quickly going to move to multimodal where it's talking to you while creating images in real time. And people, people haven't experienced that before because no human call center operator can create an image or pull up thousands of pictures in real time. But the AI can do it very easily.
- [62:03]
A
All right, let's move on if we can. A lot to cover still. We're still in AI. We're going to be covering a lot more in energy, health and starship. Nvidia beats revenues predictions defying fears of an AI bubble. I think that's great news. You know, up 56% year on year from 2024 companies at 4 trillion. Stock is up 700% since ChatGPT's 2022 release. How awesome is that? Yet we still have a bunch of US China turbulence. Nvidia gave 15% of China sales to the US to keep exporting. You know, just, you know, reporting this news. Nvidia continues to be, you know, leading the pack. There's another piece of news I want to hit on regarding this Alex I'd like you to chat about which is, which is out today and it's investors bet on cameracon as China's next AI chip champion. Would you chat about this?
- [63:08]
C
Yeah, maybe. Looking at these two stories together through the lens of where value is accumulating in the stack. So I think there are sort of two competing worldviews. One is call it the pyramid model, where the broadest part of the pyramid is at the base. In this case, under this worldview, most of the profits in the AI revolution that we're living through will accumulate at the lower infrastructure levels like the chip designers or the fabs or data centers, the lower levels. There's also competing worldview that we ultimately move to or maybe are living in, but just don't realize it yet. An inverted pyramid model where most of the profits and most of the value accrue at the upper layers, the application layers. All the startups that are being built on top of these frontier models and the frontier models themselves just become profitless or profit sucking commodities. I think if you sort of look at these two stories through a common lens at the moment, these would seem to bias me at least in the direction of thinking that for the moment, most of the profit is accumulating at the bottom of the stack at the chip design level, at the data center level, regardless of geography.
- [64:27]
A
But let's talk a little bit about this new company, about cameracon, if you would.
- [64:33]
C
Yeah, no, it's difficult to know what precisely is going on inside any given company, regardless of whether it's US based or China based. I do think again, just generalizing over Huawei, campbellcon and then obviously a whole cohort of American AI chip designers. I think we're seeing the beginnings of a non monoculture where there are diverse chip architectures, diverse chip architectures for AI acceleration from the US and seemingly the beginnings of a diverse set of non Nvidia based AI accelerator or accelerated compute architectures coming out of China. And where all of this goes, I think Gitter, your bet is as good, if not better than mine. But I think that the headline here is that there may be the beginnings of a post Nvidia, post cuda monoculture kind of China.
- [65:33]
A
I think that's the point I want to make. Whenever you restrict China on ability to sell them products, they will develop products there. We have a lead. That lead is getting shorter and shorter on chips and AI. We saw this as well in the satellite world. Right. When the U.S. defense State Department started limiting the ability to export satellites from the US to different parts of the world, the industry finally materialized and competed back against the United States. And so this strategy of scarcity doesn't work in a global culture of innovation.
- [66:10]
B
Well, I'll also tell you, David Sacks, talking to you right now, but if you, if you look at that 7 nanometer capability and we're operating at 2 nanometers, you're like, oh, we're miles ahead of China. But the algorithmic improvements can be massive. Like ten hundred thousand x kind of improvements. Yes, that are way more important than the 7 versus 2 nanometer gap and that we're not used to that in government because we're used to like the nuclear arms race or the space race, where you're not going to get a 10x advantage by magic. You know, there's no, there's no, there's no rocket fuel that you can throw in there. That's one tenth the weight of the competing. It just doesn't exist. Software, it does exist. In fact, it's common. It's everywhere. And so it's very easy to get complacent about where you stand.
- [66:55]
A
In fact, when you restrict, you cause innovation in different areas here. And you're right, algorithmic, we're going to see 100x200x1000x improvements there over the Next few years. And just remember, China has won the, you know, the Math Olympic Olympiad. Now, year after year after year, they have incredible talent. And 50% of Meta's AI staff is Chinese. Let's not fool ourselves. The intelligence is there to innovate as well as here, though it's different. Here, of course, we mentioned earlier is sort of the risk capital, the entrepreneurial drive that has people working around the clock. I would just make this note again as we talk about Nvidia. This is not going to be solely in Nvidia world. We're going to see China step up and, and compete. I love this article. Again, we've talked about the idea that that AI is no longer US centric. We're seeing the world step up and get involved. So this is billionaire Ambani taps Google and Meta to build India's AI backbone. Mukesh Ambani launches Reliance Intelligence Ventures to build India's AI infrastructure. And you know, I know Mukesh. I've been to his home a number of times in India. I was at his epic wedding, what was a year and a half ago or so. And the guy is an incredible entrepreneur. I mean, just for people to understand this. So he enters India's telecom market in 2016. He's the 10th mobile provider, right? You've got Vodafone, Airtel, all the players there. But he comes in with a completely different business model and that's his brilliance. So RelianceGeo launched a radically different model. Free voice calls for life, ultra cheap data and months of free service trial. The other thing he did was it used to take you like two days to get a mobile phone. He basically said, show up in the store, sign a few papers and it's instantly up and operating. And then he uses his capital to leapfrog over 2G and 3G and build out a 4G nationwide network. And so he literally destroyed the competition. And they are the major cell phone provider, mobile phone, telephony provider, and Saleem's in India right now. When I was there, it's 5 bar service, 5G across the nation, everywhere. So it's pretty extraordinary. And I expect he's going to do the same thing here in AI.
- [69:36]
B
Well, something big is going to happen in India because you saw Kevin Wheels saying they're making a huge push at OpenAI into India. Like, oh, that's kind of odd. Why are you doing that? Well, if you look at the demographics of the country, it has by far the most untapped talent in the world. I mean, by far China's in a terrible spot because of the aging demographic problem. The one child per family thing caught up to them in a big way. And now they've got a massive aging demographic problem. US is in great shape because immigration is strong, always has been, hopefully always will be. But India has the best latent right in the sweet spot, you know, 20 to 40 year old talent pool in the world. And so, you know, the reason that per capita GDP has been so bad in India for so long is that it's incredibly corrupt. All the, all the structures are terrible, but I think AI might have a way to cut through that and just go direct to the people.
- [70:30]
A
And also poor transportation between, you know, the roads being flooded out. We're going to see aerial delivery in India as well. All right, let's keep moving on time. Top 100 AI for 2025. So this was their, their issue. I sent Mark Benioff. Congratulations. I said, mark, you're not listed here, but you need to be on this list as well. But check this out. You know, if these are in order, Matthew Prince is number one, Elon's number two, Sam Altman's number three. And it's fascinating that Matthew Prince is number one. Any idea why?
- [71:07]
C
I mean, I would be remiss if I didn't note that sometime in the media cycle over the past week is a lot of interest in the future of Cloudflare, which Matt leads, and agentic AI. There's a lot of interest in what does almost, what does a web where AI agents that are independently surfing the web on the same level with the same rights as human web surfers look like, or should there be sort of a separate entrance to the web and to the economy for AI agents? So if I had to speculate, I would say the intersection of Cloudflare and special handling of AI agents could be one possible reason.
- [71:56]
A
I did a little digging. Let me tell you what I found out. So Matthew Prince stands atop this list for one reason. He's been focused on safeguarding the value of Internet content. So he's all about making sure that there's proper attribution and that you basically are not stealing from the publishers. And of course, Time magazine is a publisher. And so I think they're flexing their muscle here to say attribution is critically important.
- [72:28]
C
I think it's going to be, I mean, maybe even worthy of much more dedicated time actually doing a deep dive on the issue of should AI surfing the web on its own be treated the same as a human web surfer, or should they be treated differently. I think there are so many nuances there.
- [72:45]
A
The other thing we see on this list is a huge amount of global diversity and it picks up leaders in different countries. This is no longer just a Silicon Valley play, this is a global play where countries are beginning to invest heavily and really double down. On this next topic here. I love this one. So Airbnb's co founder Joe Gebbia, who's been on my stage at Abundance, he's amazing, is named the US Chief Design Officer so appointed by Trump. His goal is redesign government sites and services to be simple, modern and friendly. I love his quote. I want to make government services as satisfying to use as the Apple Store. That would be awesome.
- [73:31]
C
So I think it's perhaps not obvious, but there is actually an open source Library hosted on GitHub that I think offers Joe enormous amounts of leverage for the task that, that he's taking on. It's called the US Web Design system, US wds. And it is in principle common set of user interface components underlying most, not all perhaps, but most US government websites. And that's sort of a seminal place. I think Joe has the opportunity, such high leverage to start if the goal is to radically improve the user experience, interacting at least digitally.
- [74:09]
A
I think the key, the key point of this story here is the Trump administration tapping entrepreneurs to come in and help move the government forward. You know, despite, you know, if you're a Trump lover or hater, doesn't matter. This is about bringing in the smartest people. Because historically going to work for the government was not what where a intelligent entrepreneur would go. And there's been an incredible shift in that regard. Dave.
- [74:39]
B
Yeah, no, you phrased it exactly right. I think when the US Government said we're going to have a chief technology officer back under Obama, Originally the first two CTROs in the United States had law degrees and they were just buddies of the president. And then we created that healthcare.gov site. It was a billion dollars to build a website and then it never launched, it failed. So like, okay, why don't we get some real technologists into dc? I can't believe it's actually happening though. It's amazing.
- [75:09]
A
Well, these people are post abundance themselves, right? They've made their money. They could be working on their next moonshot or they could be building a moonshot that will hopefully. Right. The battleship or. I don't want to use that term. Right. The ocean liner of the United States.
- [75:28]
B
I think during COVID a lot of people who normally didn't care about government suddenly started Carrying a lot. They realized how much government can change your day to day life. You know, forcing you to, to stay inside. That's pretty extreme in terms of government intervention in day to day life. So, you know, whether it was right or wrong, they, they, they felt like, wow, this really matters. I need to get involved.
- [75:49]
A
Yeah. Well, good luck to Joe. I am sure. I mean this will have a huge impact. I mean, making something actually usable. This is like when we had Arpanet, you know, usable by a few individuals at MIT and Harvard and Stanford and you know, defense industry. And then Marc Andreessen comes and builds a layer on top of that with Mosaic. So if Joe can do that, make it easy to use and functional, that would be amazing. All right, here's our debate and discussion for today. I'm going to read this out and I want to hear your thoughts here. So here it is. Will people vote AI to power? So this is a tweet from Vrazerx. Because they've tired of corruption and broken promises, AI will provide laws without loopholes and policies based on measurable outcomes. Election by election, trust will shift. Eventually the ballot will include a new option. Governance by AI. Citizens will choose it not out of fear, but hope for fairness. Power won't be inherited or bought. It'll be optimized and accountable. Democracy. Democracy's paradox. People will freely vote to be governed by something beyond human flaws. So here's the question. Do you, do you believe this? Do you believe that we will be voting AI into power? Dave, what's your position here?
- [77:14]
B
Well, I think there's a long history of laws having people's names on them, like, you know, the Obamacare or Glass Steagall act or Graham Dodd or, you know, people. I think that the fact that AI is coming up with the idea and writing the law won't change the fact that some, someone will put their name on it and say, this is, this is my act. But it'll still be AI creating the law under the covers. I think it's inevitable it's going to happen. It's going to happen very, very quickly because the number of things that need some kind of a framework is explosively growing, exponentially growing. And so the traditional process of pass it through Congress, pass it through your local legislature, it's way too slow to keep up with the rate of change. So this is definitely going to happen, but not quite the way you're not going to vote in AI to be your politician. It'll still look and sound like a person.
- [78:05]
A
So I'm Going to take the. Well, let me be clear. I wish this would happen. I'd love to see this happen. I don't think there's any way in the world short of a revolution or starting a new country off world that we're going to see this happen. And there's lots of reasons. I mean, for me, the most important thing is the entrenched bureaucracies, right? Politicians, bureaucracies, entrenched interests will fiercely resist bringing this on. Courts will strike this down and talk about getting rid of corruption. Corruption doesn't vanish, it just shifts. Right? So corruption will shift from the politicians to the engineers or corporations or states that are manipulating the AI. So as much as I'd love to see this happen, I don't think it will. Alex, how about you?
- [78:53]
C
I'll take a third position in this debate.
- [78:55]
A
Of course you will.
- [78:57]
C
Nonsensical. This is a very old trope in fiction. So just two examples. If you remember the original version of the Day the Earth Stood still based on the sci fi novella Farewell to the Master, the entire premise was that alien civilizations had decided that they themselves, the biologicals, couldn't be trusted to maintain peace, so they ceded all authority to race of robots that police them. One can look back even further. Remember famously Henry vi, let's kill all the lawyers. This is a very old trope in fiction. I think it's completely nonsensical. What I expect to happen is humans will merge with the AIs. And so the question then degenerates to will people vote people to power? And the answer is yes. But it's sort of vacuous in my mind to ask whether people will be separately from that voting AI to power.
- [79:55]
A
We will couple and we will. Yeah, we'll merge, we'll speciate. All right, so next article on our economy. Jensen announces that he expects $600 billion a year on AI alone. We're seeing a massive continuation of investment. This is a good thing. We're also seeing the AI spend frenzy is propping up the US real economy and we've seen the impact surging our GDP 1%. I think this is interesting that AI infrastructure will reach 375 billion by the end of this year and is expected to be at a half a trillion in 2026. So the money is flowing out of Silicon Valley, out of sovereigns, out of family offices, into the US economy through the piping of AI. I want to pause on this conversation here. This is NASDAQ bubble soaring past.com records. So I'll read this Here's a chart here looking at the Nasdaq market over the last 25 years. So Nasdaq's market value has surged to an unprecedented levels now equal to 176% of the entire US money supply, 129% of the GDP. Both ratios are far above the dot com bubble peak signaling stock prices are racing far ahead of the real economy. Let's talk about this for a second. I think it's important. Is this different than the dot com bubble? Alex, any thoughts?
- [81:27]
C
I want to pose a thought experiment. So if we were on the verge of artificial superintelligence, what would you, Peter? Dave, expect the ratio of the NASDAQ market cap to the M2 to look.
- [81:39]
A
Like approaching infinity, ripping upward?
- [81:43]
C
Yeah, exactly. So this has all the hallmarks of the signature that one might expect, at least in the short term. One might reasonably expect there to be a concentration of rents around key publicly traded on the Nasdaq providers of AI infrastructure. And then maybe at some point, again this is not investment advice, this is idle thought experiment. One might expect perhaps at some point all of these rents become more evenly and profits become diffused throughout the economy and then maybe we see a plateau at that point. But this is exactly the signature that I would expect to see this ratio to M2 ripping upwards in the context of the eve of superintelligence.
- [82:29]
B
Yeah, I completely agree Alex. It is exactly what you would expect to see. I also don't think the dot com bubble was really a bubble in the sense that at the peak there Amazon was probably a bargain and then it went down 90 plus percent in the trough and then we had 9, 11 right after that, which turned out to be a great, great buying opportunity in the market. But the Internet was real, it was always real. And the valuations they got very high. But some great companies were in there and then Google got started right at the bottom and then went public in 04 on this chart. And so I think there's a possibility of the market coming down through panic, but it's not rational because what's going on should drive this to the moon.
- [83:16]
A
We also have companies that are real, that are profitable, that have real products and real services that are very different from the dot com world, from, you know, petfood.com days and errors. So.
- [83:28]
B
Well, a lot of these charts are meant to, meant to scare you too because here you're looking at the Nasdaq. Well, the NASDAQ is a bigger fraction of the market now because tech has become so big and the PDE is a little high of the market as a whole right now, but it's not nearly as outlandish as this chart makes it look.
- [83:45]
A
I like Alex's explanation the most. This is a signal that digital superintelligence is arriving. Let's move on to a conversation around health. One of my favorite subjects. We just saw an announcement out of the UK of an AI stethoscope detecting major heart disease in 15 seconds. This is a perfect use of technology. You put the AI layer right there at the stethoscope because, you know, in medical school, you are listening carefully to all of these heart sounds and trying to hear a murmur and trying to hear, you know, lub dub and variations thereof. AI can pick it up far, far better than this can. You know, we had a $10 million Qualcomm Tricorder X Prize. Of course, everything comes back to Star Trek, Alex, doesn't it? You know, reinventing or making the Star Trek universe real.
- [84:44]
C
Such a strange universe, Peter. Again, biotech without longevity. Very strange universe.
- [84:50]
A
Yeah, we'll get to that in a minute. But we had this $10 million competition that Paul Jacobs, who was CEO of Qualcomm, funded at the time. And to win. To win this competition, you basically had to diagnose 13 different conditions from, you know, anemia, diabetes, pneumonia, sleep apnea. The device had to weigh under five pounds, which is huge. Eventually, these things will become embedded, and you had to record five vital signs. So this is a step in the right direction. But, you know, this is great. It's still the beginning, but it portends what's coming next. And one of the things I love, one of the companies I venture back to through Bold, is called Echo Exo, and they build an ultrasound platform. But what was great about this ultrasound machine, think about the kind of device that's a handheld ultrasound that you can look at your baby or look at your carotid artery and so forth. But the key was it had an AI layer that would direct you on where to move the probe. So it would say, can you move it upwards? Can you rotate it inwards? Can you hold it there longer? And so if you had this ultrasound probe, you became the physician. The AI guided you to do what you needed to do, and then it analyzed the imagery and gave you a diagnostic. I think that's pretty amazing stuff.
- [86:18]
C
Totally. And I would also maybe invoke the statistical folklore that everything is correlated. So I think this is just scratching the tip of what's possible in principle, going back to Star Trek tricorder The key scenario I would like to see and would hope to see unlocked is mostly using the power of AI using relatively de minimis hardware. Can we simply infer the physiological state of an entire person at a distance from a few key at a distance biomarkers with AI?
- [86:49]
A
Well, I mean this is AI is going to drive healthcare out of the doctor's office, out of the hospital, into the home, right? Where you're being sensed all the time and your AI agents are just watching and listening. And it's going to transform health. And all of this will be cheap and free. It's going to be free because your company or your insurance company is going to pay for you to have those sensors in your home, on your body, in your toilet, because it just saves all the cost, right? It's like healthcare insurance is about keeping you healthy, not paying you after you've been sick. So anyway. All right, here's another fun article. I first saw this from David Sinclair who posted it. Psilocybin shows striking anti aging effects in old mice. And I added this because we have a community of folks out there who are interested in the psychoactive molecules. And so check this out. This is a 2025 study that extends cell lifespan by up to 57%. And so in this study they took aging mice and you can see on this chart here that about 20 weeks in, and this is sort of like late middle age, they started dosing them with psilocybin. And at the end of the experiment, which is 28 weeks long, and mice really just lived two years, typically to two and a half years, the survival rate of those on psilocybin was 80% versus 50% for those who are not on psilocybin. So you know, I'm trying to find out what the dose equivalent in for humans are, but just this kind of continuous discovery of different molecules that impact longevity. So those. Yeah, go ahead.
- [88:36]
C
And maybe just to comment, Peter, on this one, I mean, sorry, I read the paper, very interesting paper. I think it's potentially promising. The authors do, I think a great job of extracting downstream impacts. So not just, I think they were dosing with psilocin, which is a metabolite that normally in humans and other large mammals emerges from metabolizing psilocybin. But they look at the downstream impacts. So there's cert one expression, there are changes to the way telomeres in chromosomes are managed and regulated. That's a template, I think, where ideally one wants all of the anti aging effects without all of the central nervous system Psychedelic effects. And I think ideally. And so I think that in an ideal world we find that we were able to extract that template, CNS effects notwithstanding, we subtract those out and we're able to distill out a template for a non CNS version of this. I think that would be enormously impactful.
- [89:38]
A
Yeah. I wanted to share something I did on my summer vacation with our viewers and readers. It was something, you know, I'm always experimenting, always, hopefully doing intelligent experimentation. But when I see technology come along that I believe has a pro longevity, high reward, lower risk approach, I'm open to trying it and researching it and then sharing the results. So a couple weeks ago I posted this on X. I went and did something called stem cell re education. This was the work under Dr. Zhao. I just want to share it because for a certain group of people this will be transformational. So stem cell re education. For about six hours or so I was my blood supply, which is typically four liters, I mean sort of, I'm sorry, five liters was put through a machine two and a half times. So it was a total of 12 liters of blood were cycled and my immune cells, right, my T cells, macrophage, lymphocytes and so forth were extracted out of that. And I filled up a bag of about 300cc, a third of a liter of my white cells. Those cells were then co incubated overnight with cord blood stem cells. These are stem cells from a newborn. And those effectively my immune cells went to school. They were, they were put to a factory reset and they were, that would happen for about a 24 hour period. And the next day I had what, about 1.27 billion re educated immune cells flowed back into my body. And my goal is to bring my immune levels, my immune system levels back to a much more youthful state, reduce inflammation, rebalance my immune system, actually pump up my stem cell functionality and increase my immune function. That's my goal for myself. And we're going to be flowing this technology in through Fountain Life as well. Our goal is to set this up at our Florida centers. But I want to share, so it was amazing. Kudos to Dr. Zhao who has pioneered this work at Throne Bio. But I want to show two remarkable videos. If someone in your life is dealing with type 1 diabetes, with alopecia, with Parkinson's, with ALS, this technology is life saving. And what we're doing here, a lot of these diseases turn out to be your immune system attacking your own body, right? Alopecia, the loss of hair, that's attacking the hair follicles throughout your body and you lose your hair supply. This process basically cures that. You regrow all of your hair. Let's take a look at two videos I'm going to show you. First, a 17 year old teenager who has type 1 diabetes.
- [92:41]
C
Right.
- [92:41]
A
This is where immune system is attacking your eyelid cells in your pancreas and you no longer producing insulin. And this young man has developed a neuropathy. And you can see him here. Prior to treatment, he cannot get out of his bed to get into a wheelchair. That's his normal state of function. Now let's take a look at two months later after the treatment. I mean, it's a resurrection. He has been able to regain his function. His type 1 diabetes has been eliminated. It's extraordinary. Equally extraordinary. This is a Fountain Life patient who has als. ALS is a death sentence for most individuals. And so I want to show you the pre and post. So this is the pre video, his inability to raise his hands above his shoulder. Right. This is massively difficult. Now let's look at a couple of days later and his ability to basically regain his function for someone with als. I wish we had this for Stephen Hawking while he was still alive. So I don't know if you want to comment on this, but I just wanted to share because I think it's, you know, this is the kind of regenerative medicine that our healthspan revolution is undergoing right now.
- [94:03]
B
Well, the only comment I'll make is that when you, when you talk about allergies and autoimmune disorders, there are so many interactions going on, it's immensely complex and it's a perfect fit for AI that is just specific to your body and your results. And so it's so promising that you can actually do something with this immense amount of data we can gather now.
- [94:29]
C
I think, Peter, you're courageous and now we know what you did last summer.
- [94:36]
A
I'll report on the results. A huge list of markers were collected prior to my treatment and then I'll report out at 1 month, 3, 6 and 12 months. And I'm excited for it and we'll see where this goes. All right, let's dive into robots, energy and transport. I love this. We've seen Nvidia with I love their, their name of this. It's the Jetson AgX Thor generation of robot brains. Right. Enabling real time, intelligent interaction at the edge. The Jetsons are here. So this delivers 10x more than their previous chipset Orin. It runs generative and Reasoning AI models at the edge. I love this. And I double check these numbers. Two million developers are using the Jetson Thor development kit right now.
- [95:29]
C
Alex, 2. Two petaflops of floating point 4 FP4 compute. So for reference, that's approximately a tenth of a Blackwell or maybe about 30 iPhone 16 pros. So when we talk about what's going on with all the CapEx that's flowing into AI data centers, I don't think that's going to be bottled up in data centers for very long. We're going to see these AI chips, the AI compute, literally start to walk out of the data centers onto the streets of the rest of the economy, the so called real economy. And I think Thor is such an interesting case study in how this AI compute is going to be embodied in humanoid robots in an ergonomic, both energetically and physically ergonomic form factor and literally walk out onto the streets of the real economy.
- [96:23]
B
Yeah, and I really think this is important for all the, all the students struggling with their largely irrelevant curriculum. This is what you should be doing. Your phone or your laptop will do about 30 teraflops now. So it's about a hundredth of what you can get from a probably 4,000, $5,000 Nvidia chip. So get yourself an accelerator on the side or just run. If you go to Andre Karpathy's libraries on GitHub, you can get a really fast start and a lot of things that were daunting six months ago, you can just voice, code, vibe, code them to existence on your laptop in under an hour. And so you really can get your hands dirty with these toolkits. And then all of a sudden you're one of these people that's getting the $100 million signing offer. Like, how did I get from here to there? Well, I just jumped in and got my hands dirty.
- [97:10]
A
I played. It's fun.
- [97:11]
B
It really is fun.
- [97:12]
A
We got an image here of Jensen with Brett Adcock, CEO of Figure. We had Brett on stage at the Summit last year. We're going to have at least four, maybe five of the robot CEOs on stage with us at the Abundance Summit this coming March. And by the way, you're going to have the moonshot mates, all four of us. We'll have Saleem, Dave and Alex and myself on stage at the Abundance Summit as well. We'll be doing a moonshot sort of recap. A WTF just happened in tech in the last 3 days of the Abundance Summit in March. If you want more information, you can go to abundance360.com to learn more about the summit. It is for me an epic part of my year and my life. Getting ready for that. China's humanoid robot sales expected to exceed 10,000 units in 2025. Year on year growth of 125%. I love this image of robots marching down the street. What could possibly go wrong? But this is just the beginning, right? We're going to see. We heard when we interviewed the CEO of 1X Technologies, Bernd Bornick. He expects to see flowing out of his factories, hundreds of thousands. And of course we'll see that from figure. We'll see that from Tesla, the prediction of 10 billion humanoid robots by 2040. It's coming. All right, here's another article. I think this is pretty much Elon classic. Tesla is shifting Optimus training strategy to vision only. So we saw this with self driving. He said, no lidar. I introduced the CEO of Luminar to Elon at a party and Luminar makes a lidar. And Elon just went, nope, no lidar. We're vision only. If a driver, if a human driver can drive with like one eye, we should be able to have AI do the same. And so the switch here is no longer motion capture suits. It's just going to be training robots based upon video recordings of workers doing the work. Alex, you buy this?
- [99:23]
C
Yeah. I mean, obviously this does rhyme with the lidar versus non lidar episode with autonomous vehicles. But I think the real story here is to the extent that you believe that we're about to all be wearing smart glasses, that that's the next major form factor after smartphones. I can only imagine what fleet learning is going to look like when you have billions of people basically doing visual based motion capture for humanoid robots. Going back to the beginning of this pod. When we're talking about the bitter lesson, the bitterest lesson of all, arguably for humanoid robotics, is going to be when we have billions of people wearing smart glasses doing fleet learning to power every single trade, every single manual trade, just off of passively watching through the smart.
- [100:10]
A
Glasses and recording all of human history in detail at the micro level. Right.
- [100:17]
B
That did come up when we were at 1X Robotics. You know, Bert Barnick in his first 10,000 odd units, you have to use the fleet learning. There's no option to turn it off. So all the data, telemetry and everything from your household is getting transmitted to the central learning engine. So there's no human analog for that. Every job is training the centralized version. So if it knocks over a coffee cup in somebody's house, then the other 9,999 houses. The robot doesn't knock over the coffee cup.
- [100:47]
A
Here's my question to everybody listening and watching. Can you feel the acceleration? Can you feel the singularity coming? Oh my God, I can for sure. This is just in. Today, Apple is mandating all of its manufacturers, all of its tier one suppliers. Automate. Automate. Automate. Use robotics instead of humans everywhere possible to increase reliability of the product and reduce costs. Alex, a quick comment on this one.
- [101:21]
C
Yeah, I think if you think the world finds its way towards a completely redomesticated supply chains, I think robotics is probably the missing X factor for how, just as we were discussing earlier, tiling the world's surface, tiling the earth's surface with inference compute. One can imagine a not too distant future where robotics enables essentially every sovereign country to in some sense redomesticate its entire supply chain. If it has inference time, robotic capabilities to onshore every last bit of manufacturing.
- [101:58]
A
Amazing. We'll see this at Amazon, we'll see this at FedEx, we'll see this at all the companies that survive, the companies that don't do this aren't going to survive. I think it's going to be that pretty cut and dry. I'll want to hit one more robot story here, which is the competition between Waymo and Uber. It isn't really competition right now because we see Uber delivering 30 million trips per day while Waymo is at 700,000 trips per month. But here's the point of this one. Waymo Robotaxi outperforms 99% of Uber drivers on a daily basis in terms of daily trips. Right. These Waymos are efficient and they're running 24, 7 except for their charge time, of course. And so imagine as these roll out more and more they will displace Uber. Uber is trying their own autonomous play. They're doubling down in San Francisco. They'll be increasing the fleet by 50% there. They're trying to get into New York. They'll have a lot of resistance there. But we'll see these technologies and of course cyber taxi and cyber cattle coming.
- [103:09]
C
Alex, Dave, remember, Peter, at least in America and some other countries, the post World War II consumer automobile arguably created suburbia. Create suburb. What happens to urban planning when the cost of mobility is driven to zero? Do we see suburbs expand? What happens to roads? What happens to parking lots? What are we going to do with all the parking lots? That'd be great. I mean, but again, the elephant in the room, on top of all of the hand wringing that I want to have. Yeah, exactly. The hand wringing over urban planning is there are so many other changes that are going to happen probably in a much more, much faster timescale than we can replan cities and suburbs that maybe it's all meaningless.
- [103:51]
A
Anyway, my plan for the parking lots, are you turning them into vertical farms. Right. Each layer of the, of the. Of the parking lot is growing a different.
- [104:01]
C
Which is great if we need to be densely clustered together, but if we don't need to be densely clustered together, maybe it's something else entirely.
- [104:07]
A
And by the way, you know, I talk about the demonetization of everything, right? So driving will be four times cheaper than owning a car. So the poorest people will be chauffeured around first and foremost. And then how do you change the cost of living? Well, if you can live an hour from downtown la, where the real estate is cheap and you fly, you know, a midnight an Archer midnight evtol back and forth to work or you don't go to work, you're using Starlink to telecommute.
- [104:35]
C
Or we move to other planets or we upload to the cloud. There are so many different options.
- [104:41]
A
I'll take all of the above, please. All right, so I'm going to close on this particular piece, which is the US electricity spike begins. So this is a chart of the U.S. consumer Price Index for electricity. And we're beginning to see a spike that started in 2021 and is continuing. Thoughts on this, Alex?
- [105:07]
C
I mean, superficially, price signals convey demand. That's why we have a price based system. But I think the elephant in the room here is what happens if and when we get to recursive self improvement. We've already seen at least one demand shock, if you will. That was the Deep Sea Sputnik moment, if you will. What happens if and when there is some new algorithmic breakthrough that suddenly, radically reduces the compute intensivity of frontier models? Could the law of straight lines be violated and this burning upward, tearing upward of electricity costs reverse?
- [105:45]
A
Yeah.
- [105:46]
B
Remember electricity is so easy to predict because we know recursive self improvement is here right now. Well, I know it anyway. I think the world will know it very soon.
- [105:57]
A
I tweeted this last week. The human brain operates on 20 watts of energy. And I was playing in GPT5 and asking what the equivalent compute cost in terms of energy for one of the frontier models. And it's somewhere between 100,000 and a million times more energy than the human brain. So this is a massive potential for improvement here. We'll get new chip designs, new, you know, new strategies and approaches to make it more efficient. So we'll build out all of this, you know, energy data centers and then if we, you know, 10 to the fifth, 10 to the sixth improvement in energy efficiency, that means we get that level of improvement on our, our total AI capabilities.
- [106:41]
C
We can do better than that, Peter. The Landauer limit, we can blow past it with reversible computing. Human brain is by no means an optimal computer. There are lots of other better ways we could build computronium.
- [106:53]
A
Amazing. Guys, listen, as always, I love spending my time with you. I feel smarter afterwards. I hope everybody listening enjoyed this episode. A lot more coming we're going to see the release of Gemini 3. We'll be back on to discuss that. We're going to be coming on with a lot more of WTF just happened in tech. Hopefully this is your dose of optimism to counter all of the moaning pessimism coming off of these, the media channels that people normally consume. I've stopped watching the news. For me, this is the news. The news that really matters, that transforms our planet, that is giving us increased longevity. It's going to increase sustainable abundance is the word. Alex, some closing thoughts from you and.
- [107:38]
C
Then we'll go to Dave Black Hole supercomputers.
- [107:43]
B
Okay, okay, that's a closing thought which.
- [107:46]
A
Assumes we're not living in a black hole right now.
- [107:51]
B
We'll follow up on the short term implications of the Leopold list and a bunch of other things. I think we'll be back online again very quickly within a week. With Salim back, there's so much happening now. I have a bunch of things we couldn't even get to today. And then more will happen within within the week. So yeah, just a lot, a lot to keep up with, but this is the place to do it.
- [108:11]
A
Yeah, I was wearing my Occupy Mars shirt in expectation that we'll discuss Starship 10 launch which was a huge success. So congratulations to Elon and the team at SpaceX for that launch. It was awesome. If you haven't seen the video, please, it is a proof that we're living in the year 2025. Humanity is building fusion, going to Mars, heading towards longevity, escape velocity. The only time more exciting than today to be alive is tomorrow. On that note, gentlemen, have a beautiful week. Talk to you all soon.
- [108:48]
D
Every week my team and I study the top 10 technology metatrends that will transform industries over the decade ahead. I cover trends ranging from humanoid robotics, AGI and quantum computing to transport energy, longevity, and more. There's no fluff, only the most important stuff that matters that impacts our lives, our companies, and our careers. If you want me you to show share these metatrends with you, I write a newsletter twice a week, sending it out as a short 2 minute read via email. And if you want to discover the most important meta trends ten years before anyone else, this report's for you. Readers include founders and CEOs from the world's most disruptive companies and entrepreneurs building the world's most disruptive tech. It's not for you. If you don't want to be informed about what's coming, why it matters, and how you can benefit from it. The substantial subscribe for free. Go to dmandis.com metatrends to gain access to the trends 10 years before anyone else. All right, now back to this episode.
- [109:56]
B
This episode is brought to you by LifeLock.
- [109:59]
C
Between two factor authentication, strong passwords, and a VPN, you try to be in control of how your info is protected. But many other places also have it, and they might not be as careful. That's why LifeLock monitors hundreds of millions.
- [110:12]
B
Of data points a second for threats.
- [110:14]
C
If your identity is stolen, they'll fix it, guaranteed, or your money back.
- [110:19]
B
Save up to 40% your first year. Visit lifelock.com podcast for 40% off terms apply.